IMPORTANCEThe current definition and staging of acute kidney injury (AKI) considers alterations in serum creatinine (sCr) level and urinary output (UO). However, the relevance of oliguria-based criteria is disputed. OBJECTIVETo determine the contribution of oliguria, as defined by the Kidney Disease: Improving Global Outcomes (KDIGO) criteria, to AKI diagnosis, severity assessment, and short-and longterm outcomes. DESIGN, SETTING, AND PARTICIPANTS This cohort study included adult patients admitted to a multidisciplinary intensive care unit from January 1, 2010, to June 15, 2020. Patients receiving longterm dialysis and those who declined consent were excluded. Daily sCr level and hourly UO measurements along with sociodemographic characteristics and severity scores were extracted from electronic medical records. Long-term mortality was assessed by cross-referencing the database with the Swiss national death registry. The onset and severity of AKI according to the KDIGO classification was determined using UO and sCr criteria separately, and their agreement was assessed. MAIN OUTCOMES AND MEASURES Using a multivariable model accounting for baseline characteristics, severity scores, and sCr stages, the association of UO criteria with 90-day mortality was evaluated. Sensitivity analyses were conducted to assess how missing sCr, body weight, and UO values, as well as different sCr baseline definitions and imputations methods, would affect the main results. RESULTS Among the 15 620 patients included in the study (10 330 men [66.1%] with a median age of 65 [IQR, 53-75] years, a median Simplified Acute Physiology Score II score of 40.0 [IQR, 30.0-53.0], and a median follow-up of 67.0 [IQR, 34.0-100.0] months), 12 143 (77.7%) fulfilled AKI criteria.Serum creatinine and UO criteria had poor agreement on AKI diagnosis and staging (Cohen weighted κ, 0.36; 95% CI, 0.35-0.37; P < .001). Compared with the isolated use of sCr criteria, consideration of UO criteria enabled identification of AKI in 5630 patients (36.0%). Those patients had a higher 90-day mortality than patients without AKI (724 of 5608 [12.9%] vs 288 of 3462 [8.3%]; P < .001).On multivariable analysis accounting for sCr stage, comorbidities, and illness severity, UO stages 2 and 3 were associated with a higher 90-day mortality (odds ratios, 2.4 [95% CI, 1.6-3.8; P < .001] and 6.2 [95% CI, 3.7-10.5; P < .001], respectively). These results remained significant in all sensitivity analyses. CONCLUSIONS AND RELEVANCEThe findings of this cohort study suggest that oliguria lasting more than 12 hours (KDIGO stage 2 or 3) has major AKI diagnostic implications and is associated with outcomes irrespective of sCr elevations.

Introduction: Regional citrate anticoagulation (RCA) is the recommended anticoagulation modality for continuous renal replacement therapy (CRRT). RCA was associated with a low rate of complications in randomized controlled trials. However, little is known about the type and rate of complications in real life. We sought to describe complications associated with RCA in comparison with those associated with heparin anticoagulation. Methods: In our institution, RCA has been the default anticoagulation modality for CRRT in all patients without contraindications since 2013. We have retrospectively reviewed all consecutive patients who received CRRT between January and December 2016 in our institution. For each CRRT session, we have assessed circuit duration, administered dose, as well as therapy-associated complications. Those parameters were compared according to whether the circuit was run in continuous veno-venous hemodialysis (CVVHD) mode with RCA or continuous veno-venous hemofiltration (CVVH) mode with heparin anticoagulation. Results: We analyzed 691 CRRT sessions in 121 patients. Of those 400 (57.9%) were performed in CVVHD-RCA mode and 291 (42.1%) in CVVH-Heparin Mode. Compared with CVVH-Heparin mode, CVVHD-RCA mode was associated with a longer circuit lifespan (median duration 54.9 interquartile range [IQR 44.6] vs. 15.3 h [IQR 22.4], p < 0.0001). It was associated with a higher rate of metabolic acidosis 77 (20.2%) vs. 18 (7.2%), (p < 0.0001), alkalosis 186 (48.7%) vs. 43 (17.1%), (p = 0.0001), and hypocalcemia 96 (25.07%) vs. 26 events (10.79%), p < 0.0001. However, the majority of these alterations were of benign or moderate severity. Only one possible citrate intoxication was observed. Conclusions: CVVHD-RCA was associated with a much longer circuit life but an increased rate of minor metabolic complications, in particular acid-base derangements. Some of these complications might have been prevented by therapy adaptation. Medical and nursing staff education is of major importance in the implementation of an RCA protocol.

SUMMARY:A 33-year-old woman developed severe post-lumbar puncture headaches in the course of work-up for multiple sclerosis. Immediately after receiving treatment with intravenous caffeine, she became blind and experienced a generalized tonic-clonic seizure. Brain MR imaging then showed vasogenic parieto-occipital edema. She recovered clinically and radiologically within 72 hours. After 1 year of follow-up, there was no recurrence of symptoms or radiologic changes.

Background Hemoadsorption (HA) might mitigate the systemic inflammatory response associated with post-cardiac arrest syndrome (PCAS) and improve outcomes. Here, we investigated the feasibility, safety and efficacy of HA with CytoSorb® in cardiac arrest (CA) survivors at risk of PCAS. Methods In this pilot randomized controlled trial, we included patients admitted to our intensive care unit following CA and likely to develop PCAS: required norepinephrine (> 0.2 µg/kg/min), and/or had serum lactate > 6 mmol/l and/or a time-to-return of spontaneous circulation (ROSC) > 25 min. Those requiring ECMO or renal replacement therapy were excluded. Eligible patients were randomly allocated to either receive standard of care (SOC) or SOC plus HA. Hemoadsorption was performed as stand-alone therapy for 24 h, using CytoSorb® and regional heparin–protamine anticoagulation. We collected feasibility, safety and clinical data as well as serial plasma cytokines levels within 72 h of randomization. Results We enrolled 21 patients, of whom 16 (76%) had out-of-hospital CA. Median (IQR) time-to-ROSC was 30 (20, 45) minutes. Ten were assigned to the HA group and 11 to the SOC group. Hemoadsorption was initiated in all patients allocated to the HA group within 18 (11, 23) h of ICU admission and conducted for a median duration of 21 (14, 24) h. The intervention was well tolerated except for a trend for a higher rate of aPTT elevation (5 (50%) vs 2 (18%) p = 0.18) and mild (100–150 G/L) thrombocytopenia at day 1 (5 (50%) vs 2 (18%) p = 0.18). Interleukin (IL)-6 plasma levels at randomization were low (< 100 pg/mL) in 10 (48%) patients and elevated (> 1000 pg/mL) in 6 (29%). The median relative reduction in IL-6 at 48 h was 75% (60, 94) in the HA group versus 5% (− 47, 70) in the SOC group (p = 0.06). Conclusions In CA survivors at risk of PCAS, HA was feasible, safe and was associated with a nonsignificant reduction in cytokine plasma levels. Future trials are needed to further define the role of HA after CA. Those studies should include cytokine assessment to enrich the study population. Trial registration: NCT03523039, registered 14 May 2018.